Qwen3.7 Plus
Mistral Medium 3.5
- Context
- 200Ktokens
- Released
- 1y ago
Claude Sonnet 4 is an Anthropic model that succeeds Claude Sonnet 3.7 as part of the Claude 4 generation, introduced alongside Claude Opus 4. Anthropic presents it as a major upgrade over its predecessor, claiming better coding and reasoning, more precise instruction-following, and that both Claude 4 models are 65% less likely than Sonnet 3.7 to take shortcuts on the agentic tasks most susceptible to them. It is built for coding, advanced reasoning, and AI agents, and runs in two modes, near-instant responses and extended thinking.
- Multimodal model
- Accepts images, files
- Reasoning
- Tools
How it compares
Level
Mid-tierThe lab's everyday default: strong on most tasks at a mid-range price.
Vs. your stack
Add the models you use to see how this one compares.
Sign inAdditional source evidence
- Source model
- anthropic/claude-sonnet-4-20250514
- Source family
- claude-sonnet
- Last updated
- 2025-05-22
- Knowledge cutoff
- 2025-03-31
- Model context
- 200K
- Model output limit
- 64K
- Open weights
- No
- File attachments
- Yes
- Reasoning
- Yes
- Tool calling
- Yes
- Temperature control
- Yes
Published benchmarks
| Benchmark | Score | Evaluation details | Evidence |
|---|---|---|---|
| Aider Polyglot | 61.3 · percent correct | 2025-05-24 | Source |
Complete source record (JSON)
{
"id": "anthropic/claude-sonnet-4-20250514",
"name": "Claude Sonnet 4",
"description": "Balanced Claude model for coding, analysis, agent workflows, and cost control",
"family": "claude-sonnet",
"last_updated": "2025-05-22",
"knowledge": "2025-03-31",
"attachment": true,
"reasoning": true,
"tool_call": true,
"temperature": true,
"modalities": {
"input": [
"text",
"image",
"pdf"
],
"output": [
"text"
]
},
"limit": {
"context": 200000,
"output": 64000
},
"benchmarks": [
{
"name": "Aider Polyglot",
"score": 61.3,
"source": "https://aider.chat/docs/leaderboards/",
"metric": "percent correct",
"date": "2025-05-24"
}
],
"release_date": "2025-05-22",
"open_weights": false
}Text provenance
TransitioningClaude text watermarking
Anthropic says Claude models launched before August 2, 2026 have a transition period and are being moved to the same watermarking system over time.
Official explanationFrequently asked questions
Capabilities and specs
Claude Sonnet 4 has a context window of 200K tokens (200,000) and a maximum output of 64K tokens (64,000).
Yes, Claude Sonnet 4 can read images and files. Its input modalities are text, image and file. Audio and video are not listed as supported.
Yes, Claude Sonnet 4 supports tool calling (function calling).
Yes, Claude Sonnet 4 supports a reasoning mode.
The knowledge cutoff of Claude Sonnet 4 is 2025-03-31.
Quality and benchmarks
On Arena Vision, Claude Sonnet 4 ranks 90 with a score of 1206. On Arena Text, it ranks 165 with a score of 1402.
Yes, Claude Sonnet 4 is ahead of its predecessor Claude Sonnet 3.7 on ARC-AGI-2, Arena Text, OSWorld and OTIS Mock AIME 2024-2025. It is behind on GPQA diamond and Humanity's Last Exam, and level on Arena Vision and FrontierMath-2025-02-28-Private.
The closest alternatives, matched on model kind and release date (no comparable price), are Apple On-Device Model (2025) by Apple, Apple Server Model (2025) by Apple and ERNIE 4.5 Turbo by Baidu.
Apple On-Device Model (2025)Claude Sonnet 4 vs Apple On-Device Model (2025)Apple Server Model (2025)Claude Sonnet 4 vs Apple Server Model (2025)ERNIE 4.5 TurboClaude Sonnet 4 vs ERNIE 4.5 Turbo
Claude Sonnet 4 is closed, so it has no open-weight version. The highest-ranked open-weight model on Arena Text is Kimi K3 at rank 19.
Licensing and access
No, Claude Sonnet 4 is closed.
Basics
Claude Sonnet 4 was made by Anthropic and released on May 22, 2025.
Generated from Vertical Tabs data, updated November 11, 2026.
Discussion
No comments yet
Be the first to share your thoughts.